Unmanned aerial vehicle inspection system for fishing-light complementary scene

Abstract

The invention discloses a fishing light complementation scene unmanned aerial vehicle inspection system, and relates to the technical field of unmanned aerial vehicle inspection, the fishing light complementation scene unmanned aerial vehicle inspection system comprises a ground management and control platform and a monitoring device which communicate with each other through a network, and the ground management and control platform is used for regularly collecting and identifying fishing light complementation scene image information and environment data; the image information and the environment data are processed, analyzed and pre-warned, the monitoring device is used for comprehensively monitoring the image and the environment data of the fishing-light complementary scene, and the ground management and control platform comprises a region division unit, a data analysis unit, a processor, a memory and a display panel. The monitoring device comprises a ground monitoring station and unmanned aerial vehicle equipment. The system can reduce the human resource cost, can cover the front and back of a photovoltaic panel, the center and the edge of a culture area and other traditional blind areas through the mutual cooperation of ground station fixed monitoring and unmanned aerial vehicle maneuvering inspection, can cover a large-area fishing-light complementary scene, and can greatly reduce the manual inspection cost and the equipment maintenance cost for a long time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle inspection, in particular to an unmanned aerial vehicle inspection system for fish-light complementary scenes. BACKGROUND

[0002] Fish-light complementation refers to the combination of aquaculture and photovoltaic power generation, i.e., installing photovoltaic components on fish ponds, and carrying out aquaculture in the water area below the photovoltaic components, thus forming a new power generation mode of "power generation on the top and fish culture on the bottom". During the use of photovoltaic power stations, photovoltaic panels are exposed to the external environment for a long time, which makes the photovoltaic panels prone to damage, hot spots, dust blockage, and bracket corrosion. Therefore, it is necessary to conduct regular manual monitoring to timely grasp the use of photovoltaic panels and the activities of fish groups in fish-light complementary scenes. However, this monitoring method requires a large amount of manpower, material resources, and time, and has a limited monitoring range, which is easily restricted by the environment and cannot achieve real-time and comprehensive monitoring of large-area fish-light complementary scenes, thus failing to comprehensively evaluate the comprehensive benefits of the "fish-light complementary" project.

[0003] Therefore, the present application provides an unmanned aerial vehicle inspection system for fish-light complementary scenes to eliminate the drawbacks of the prior art. SUMMARY

[0004] The present application aims to provide an unmanned aerial vehicle inspection system for fish-light complementary scenes to solve the problems of manual inspection in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions: The unmanned aerial vehicle inspection system for fish-light complementary scenes comprises a ground control platform and a monitoring device which communicate with each other through a network. The ground control platform is used for regularly collecting and identifying image information and environmental data of fish-light complementary scenes, and processing, analyzing, and warning the image information and environmental data. The monitoring device is used for comprehensively monitoring the images and environmental data of fish-light complementary scenes. The ground control platform comprises: a region division unit which is used for obtaining environmental data of a fish-light complementary scene to be inspected and dividing the region to be inspected into several regions, wherein the region to be inspected comprises a photovoltaic region and an aquaculture region; a data analysis unit which is used for analyzing the collected data from the monitoring device and making real-time update prediction of the life cycle of photovoltaic panels based on the collected data; a processor which is used for coordinating the data interaction between modules; a storage which is used for collecting, arranging, and storing various data from the Internet, the monitoring device, and the data analysis unit; a display panel which is used for real-time display of the surface condition of photovoltaic panels and data information; The monitoring device comprises: The ground monitoring station is used to equip various types of sensors to collect images and environmental parameters of a specific area in real time. The unmanned aerial vehicle device is used to obtain image data of the surface of the photovoltaic panel through multi-angle shooting and collect overall environmental data in the process of fish-light complementary scene use in real time.

[0006] Preferably, the ground monitoring station comprises: The image monitor is used to shoot images of the surface of the photovoltaic panel and the surrounding area of the breeding area at regular intervals and transmit the collected image information to the data analysis unit. The temperature and humidity sensor is used to collect the temperature and humidity of the fish-light complementary scene in real time. The water quality sensor is used to detect the water quality of the breeding area. The weather sensor is used to collect air temperature, climate, light duration and intensity, wind speed, wind direction and rainfall data in the fish-light complementary scene in real time, and the weather sensor comprises a light intensity sensor, a wind speed and direction sensor and a rainfall sensor.

[0007] Preferably, the unmanned aerial vehicle device comprises a plurality of unmanned aerial vehicles and an unmanned aerial vehicle processor fixedly installed in the unmanned aerial vehicle, and the unmanned aerial vehicle processor comprises: The unmanned aerial vehicle control terminal is electrically connected with the processor and is used to receive the unmanned aerial vehicle parameters, region coordinate information and control instructions output by the processor and adjust the operation of the corresponding module according to the control instructions. The region image acquisition module comprises an infrared camera and a hyperspectral camera, the infrared camera is used to obtain image data and temperature information of the region to be inspected, and the hyperspectral camera is used to obtain the material state of the photovoltaic panel and water quality parameters. The GPS tracking and positioning module is used to track and record the position of the unmanned aerial vehicle in flight in real time and correspond to the boundary position coordinates of the region to be inspected in the storage. The obstacle avoidance and distance measurement module is used to identify obstacles on the flight path and automatically adjust the flight path of the unmanned aerial vehicle to keep a safe distance between the unmanned aerial vehicle and the obstacles in the flight process. The communication module is used to realize real-time contact between the ground control platform and the unmanned aerial vehicle and transmit the data received by the unmanned aerial vehicle control terminal to the data analysis unit. The unmanned aerial vehicle control terminal, the region image acquisition module, the GPS tracking and positioning module, the obstacle avoidance and distance measurement module and the communication module communicate with each other through a network.

[0008] Preferably, the region division unit comprises: The land division module is used for importing a region drawing to be inspected and field measurement data, establishing a region coordinate system, dividing a fish-light complementary scene into a photovoltaic area and a breeding area, and dividing the photovoltaic area and the breeding area into a plurality of sub-areas, and determining region coordinate information; The region marking module is used for recording positions and ranges of different sub-areas in the region coordinate system, and displaying whether corresponding regions have region abnormal phenomena through different colors. The path planning module is used for planning a plurality of unmanned aerial vehicle aerial photography routes according to sub-area division results, GPS positioning data, photovoltaic panel sizes and installation angles, and ground monitoring station distribution positions.

[0009] Preferably, the data analysis unit comprises: The data preprocessing module is used for performing preprocessing operations, including cleaning, denoising, format conversion and standardization processing, on collected various types of original data. The feature extraction module is used for extracting key features related to photovoltaic panel states, breeding area environments and overall scenes from the preprocessed data. The model prediction module is used for constructing a prediction model based on photovoltaic panel life cycle data, historical environment parameters and operation data stored in the storage, and real-time monitoring and outputting operation conditions and prediction results of the prediction model. The early warning module is used for comparing simulation data generated by the prediction model with real-time monitoring data, judging whether an abnormality occurs, and delivering abnormal data to the ground control platform if the abnormality occurs. The result output module is used for outputting results processed by the data analysis unit in the form of charts and reports, including photovoltaic panel life cycle prediction results, abnormality analysis and environment parameter statistical data.

[0010] Preferably, the unmanned aerial vehicle flies according to a flight path and a speed preset by the path planning module under control of the unmanned aerial vehicle processor, and image data collected by the region image collection module and image data collected by the ground monitoring station form a complement.

[0011] Preferably, the data preprocessing module is further used for converting image data collected by the region image collection module into photo data with time labels, and performing merging operations on the photo data and GPS positioning data recorded by the GPS tracking positioning module at the same time.

[0012] Preferably, the feature extraction module extracts feature information of photovoltaic panel surface abnormalities, water quality abnormalities, fish school activity abnormalities, equipment faults and strangers, and the feature information of the photovoltaic panel surface abnormalities includes damage, hot spots, dust shielding and bracket rust.

[0013] Preferably, the prediction model adopts a time series prediction model based on machine learning.

[0014] Preferably, it also includes a user terminal, which is used to receive early warning information and view relevant data. The ground control platform, monitoring device and user terminal communicate with each other through a network. The relevant data includes historical inspection data, abnormal data and prediction results.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: Compared to traditional manual monitoring methods, this system can reduce human resource costs to a certain extent. By using drones and ground monitoring stations, it can gain a more comprehensive understanding of the surface condition of photovoltaic panels, increase the monitoring range, and promptly detect abnormal changes during the use of photovoltaic panels. Through the cooperation between fixed ground station monitoring and mobile drone inspections, it can cover traditional blind spots such as the front and back of photovoltaic panels and the center and edges of aquaculture areas, covering large-scale aquaculture-solar hybrid scenarios. By using infrared cameras and hyperspectral cameras for multi-angle shooting, it can improve the detection accuracy of photovoltaic panel surface anomalies, reduce errors and omissions in manual inspections, and significantly reduce manual inspection costs and equipment maintenance expenses in the long run. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system structure of the present invention.

[0017] Figure 2 This is a schematic diagram of the ground control platform of the present invention.

[0018] Figure 3 This is a schematic diagram of the monitoring device structure of the present invention.

[0019] Figure 4 This is a schematic diagram of the structure of the drone device of the present invention.

[0020] Figure label annotations: Area division unit 100, plot division module 110, area marking module 120, path planning module 130, data analysis unit 200, data preprocessing module 210, feature extraction module 220, model prediction module 230, early warning module 240, result output module 250, processor 300, memory 400, display panel 500, ground monitoring station 600, image monitor 610, temperature and humidity sensor 620, water quality sensor 630, meteorological sensor 640, UAV equipment 700, UAV control terminal 710, area image acquisition module 720, GPS tracking and positioning module 730, obstacle avoidance and ranging module 740, communication module 750. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0022] In this embodiment, as Figure 1 - Figure 4 As shown, the drone inspection system for a fish-solar hybrid scenario refers to a comprehensive utilization model where photovoltaic panels are installed above aquaculture water bodies to generate electricity above and raise fish below. This includes a photovoltaic area (power generation) and an aquaculture area (fish farming). The system comprises a ground control platform and monitoring devices that communicate with each other via a network. The ground control platform is used to periodically collect and identify image information and environmental data of the fish-solar hybrid scenario, and to process, analyze, and issue early warnings based on this information. The monitoring devices are used to conduct comprehensive monitoring of the images and environmental data of the fish-solar hybrid scenario. The monitoring devices include 600 ground monitoring stations and 700 drones. The ground control platform includes: The area division unit 100 is used to acquire environmental data of the area to be inspected in the fishery-solar complementary scenario, and to divide the area to be inspected into several areas, including the photovoltaic area and the aquaculture area. like Figure 2 As shown, it specifically includes: The plot division module 110 is used to import the map and field measurement data of the area to be inspected. It obtains a high-precision map of the area to be inspected through UAV aerial survey or satellite remote sensing, including information such as photovoltaic panel layout, fishpond boundary and water depth data, and surrounding topographic features. Based on geographic information or CAD drawings combined with GPS coordinates, it ensures that the data matches the drawings, divides the fish-solar complementary scene into a photovoltaic area for power generation and an aquaculture area for aquaculture, and divides the fishpond into feeding area, oxygenation area, etc. according to the function. It establishes a regional coordinate system, divides the fish-solar complementary scene into a photovoltaic area and an aquaculture area, and divides the photovoltaic area and the aquaculture area into several sub-areas, determines the regional coordinate information, divides the power generation area based on the photovoltaic panel installation coordinates, combined with tilt angle and orientation information, divides the aquaculture area according to the fishpond boundary and water depth data, and generates a regional division vector map, including the coordinate range of the photovoltaic area, the aquaculture area and the sub-areas. The region marking module 120 is used to record the position and range of different sub-regions in the region coordinate system, and to display whether there are regional anomalies in the corresponding regions through different colors. It dynamically marks the divided sub-regions and intuitively displays the abnormal status. For example, the abnormal level can be displayed according to the color intensity, or the abnormal type can be indicated according to different colors, such as red for photovoltaic panel hot spot anomaly, yellow for water quality pH anomaly, etc. The path planning module 130 is used to plan several drone aerial photography routes based on the sub-region division results, GPS positioning data, photovoltaic panel size and installation angle, and the distribution of ground monitoring stations 600. According to the regional division and equipment distribution, it plans efficient inspection paths for drones to avoid duplication or omission. The input parameters are the sub-region coordinates, photovoltaic panel parameters, and the location of ground monitoring stations 600 from the plot division module 110. The photovoltaic panel parameters include size, tilt angle, installation height, etc. The scene is divided into 1m×1m grids. The algorithm calculates the shortest path and outputs the flight path in KML or Mission Planner format, including waypoint coordinates, flight altitude, and shooting angle. For example, the photovoltaic area adopts a zigzag path, maintaining a distance of 2-3 meters from the photovoltaic panel surface. The aquaculture area adopts a gridded path, and the flight altitude can be set to 5-8 meters. The inspection density and path can be adjusted according to the actual environmental needs. The data analysis unit 200 is used to analyze the data collected from the monitoring device and to update and predict the life cycle of the photovoltaic panel in real time based on the collected data. The model prediction module 230 adopts a time series prediction model based on LSTM (Long Short-Term Memory Network). The input data includes photovoltaic panel temperature, power generation efficiency and environmental parameters. The output is the predicted value and confidence interval of the remaining life of the photovoltaic panel. The collected data includes: photovoltaic panel status data, environmental data, temperature and humidity, light intensity, wind speed, rainfall and other meteorological data, as well as water quality parameters and historical data of the aquaculture area. The photovoltaic panel status data includes temperature distribution captured by infrared camera, material status acquired by hyperspectral camera, surface damage captured by image monitor, dust obstruction, etc. The historical data includes stored photovoltaic panel life cycle records, past abnormal events and maintenance records. The analysis includes: photovoltaic panel health status detection, aquaculture area water quality monitoring and photovoltaic panel life cycle prediction based on historical data. like Figure 2 As shown, it specifically includes: The data preprocessing module 210 is used to preprocess various types of raw data collected, including cleaning, denoising, format conversion and standardization. It is used to remove duplicate, invalid or erroneous data (such as sensor outliers and blurry images collected by UAVs), use interpolation or based on adjacent data points to supplement missing values, use filtering algorithms for image data, use moving average or wavelet transform to denoise environmental sensor data, and then unify data from different sources into a standardized format, convert image data to a uniform resolution, align image data with GPS positioning data through timestamps, and associate image data and environmental parameters collected at the same time with GPS location information to form a structured record. Specifically, the data preprocessing module 210 is also used to convert the image data acquired by the regional image acquisition module 720 into photo data with time tags, and to merge the photo data with the GPS positioning data recorded by the GPS tracking and positioning module 730 at the same time. The feature extraction module 220 is used to extract key features related to the photovoltaic panel status, aquaculture area environment and overall scene from the preprocessed data. The features extracted by the feature extraction module 220 include features related to photovoltaic panel surface anomalies, water quality anomalies, fish activity anomalies, equipment failures and the appearance of strangers. Features related to photovoltaic panel surface anomalies include damage, hot spots, dust obstruction and support corrosion, which can display performance such as power generation efficiency decay rate and temperature distribution uniformity. Environmental features include cumulative sunshine duration, frequency of extreme weather events, dissolved oxygen change trend, number of times ammonia nitrogen concentration exceeds the standard, etc. Through correlation analysis, features that contribute more to the prediction target are screened out to reduce redundancy. The model prediction module 230 is used to build a prediction model based on the photovoltaic panel life cycle data, historical environmental parameters and operating data stored in the memory 400. It monitors and outputs the operation status and prediction results of the prediction model in real time. The model uses the photovoltaic panel life cycle data, historical environmental parameters and operating data stored in the memory 400 as training samples. Through machine learning algorithms, the model continuously learns the patterns and trends in the data, so that the model can combine the real-time monitored data to update and predict the life cycle of the photovoltaic panel in real time, thereby timely grasping the usage status of the photovoltaic panel and providing a scientific basis for the maintenance and replacement of the photovoltaic panel. The specific model prediction operation is as follows: The prediction model adopts a machine learning-based time series prediction model. Based on the prediction target, a suitable machine learning model is selected, combined with computer vision technology to form a hybrid prediction architecture. The basic model uses LSTM (Long Short-Term Memory Network) as the core time series model to process historical sensor data, and combines random forest regression to process structured features, judging whether the photovoltaic panels have anomalies such as hot spots or damage. The data is divided into training, validation, and test sets. The training set is used to fit the model, the validation set is used to adjust hyperparameters, and cross-validation is used to evaluate model performance. A loss function is selected, and the trained model is integrated into the model prediction module 2 of the system. In section 30, a photovoltaic (PV) panel lifecycle prediction operation is implemented. The input consists of real-time collected PV panel status (temperature, power generation efficiency), historical power generation efficiency data, and environmental data (sunlight, temperature). The input data is normalized and segmented using a sliding window to form time series samples. The input layer of the LSTM model receives the standardized time series data. The LSTM layer uses two layers with 64 neurons each, and the activation function is the tanh function to capture long-term dependencies. The dropout rate of the Dropout layer is set to 0.2 to prevent overfitting. The fully connected layer outputs the predicted remaining lifespan of the PV panel (for regression problems) and the probability of anomalies (for classification problems), and provides the confidence interval. The early warning module 240 is used to compare the simulated data generated by the prediction model with the real-time monitoring data to determine whether any abnormality has occurred. If an abnormality is found, the abnormal data is transmitted to the ground control platform. For example, if an infrared image shows that the temperature in a certain area is significantly higher than that of the surrounding area, the model, combined with historical hot spot data, determines that it is "high risk" and generates an early warning. The results output module 250 is used to output the results processed by the data analysis unit 200 in the form of charts and reports, including photovoltaic panel life cycle prediction results, abnormal situation analysis, and environmental parameter statistics. The photovoltaic panel life cycle prediction results are displayed in the form of remaining service life percentage and aging trend chart. The abnormal situation classification analysis report is divided into three levels according to the urgency: red (immediate handling), orange (handling within 48 hours), and yellow (observation). The environmental parameter statistics include the spatiotemporal distribution heat map of parameters such as temperature and irradiance, and automatically generate a maintenance order containing abnormal location coordinates and maintenance suggestions. The processor 300 is used to coordinate data interaction between modules. The processor 300 can monitor the operating status of each module and automatically start the backup channel when a communication interruption or equipment failure is detected, so as to ensure that the timing of data collection between the ground monitoring station 600 and the UAV is consistent. The memory 400 is used to collect, organize, and store various types of data from the Internet, monitoring devices, and the data analysis unit 200. Display panel 500 is used to display the surface condition and data information of photovoltaic panels in real time, and supports access from multiple platforms such as PC and mobile devices.

[0023] Among them, such as Figure 1 and Figure 3 As shown, the ground monitoring station 600 is equipped with various types of sensors to collect images and environmental parameters of a specific area in real time. Sensors are deployed at key locations. For example, in a photovoltaic area, one ground monitoring station 600 can be deployed every 200 square meters, equipped with an image monitor 610, a temperature and humidity sensor 620, an irradiance sensor, etc. In an aquaculture area, one ground monitoring station 600 can be deployed every 100 square meters, equipped with a water quality sensor 630, a meteorological sensor 640, etc., including: Image monitor 610 is used to periodically capture images of the photovoltaic panel surface and the surrounding area of ​​the aquaculture area, and transmit the collected image information to data analysis unit 200. It automatically captures images according to preset time intervals or event triggers. In addition to visible light, a near-infrared camera can be optionally equipped to detect algae reproduction or abnormal water quality. The 620 temperature and humidity sensor is used to collect real-time environmental temperature and humidity data in a fishery-solar hybrid scenario. Increased temperature can lead to decreased power generation efficiency, which in turn affects the efficiency of photovoltaic panels. Water temperature can be used to analyze and predict dissolved oxygen levels in aquaculture areas. The 630 water quality sensor is used to detect the water quality in aquaculture areas, such as dissolved oxygen, ammonia nitrogen content, pH value, and turbidity. The data monitored by the 630 water quality sensor can be linked with the aeration equipment, and the aeration equipment will be automatically started when the dissolved oxygen is lower than the threshold. The 640 weather sensor is used to collect real-time data on temperature, sunshine duration and intensity, wind speed, wind direction, and rainfall in the aquaculture-solar hybrid scenario. The sunshine intensity sensor can be used to monitor solar irradiance and identify cloudy and rainy weather, and to assess the difference between the theoretical and actual power generation of the photovoltaic panels. The wind speed and direction sensor can be used to detect wind speed, and if the wind speed exceeds the threshold, the drone operation will be suspended to ensure safety. It can also analyze wind direction to optimize the photovoltaic panel cleaning strategy. The rainfall sensor is used to predict changes in water level in the aquaculture area and to evaluate the cleaning effect of the photovoltaic panels.

[0024] Among them, such as Figure 1 , Figure 3 and Figure 4 As shown, the drone device 700 is used to capture image data of the photovoltaic panel surface from multiple angles, and to collect overall environmental data in real time during the use of the solar-fishery complementary scenario. It includes several drones and a drone processor fixedly installed inside each drone. Under the control of the drone processor, the drones perform flight inspection operations according to the flight path and speed preset by the path planning module 130. Daily scheduled inspections are possible. The image data collected by the area image acquisition module 720 complements the image data collected by the ground monitoring station 600. The drone processor includes: The UAV control terminal 710 is electrically connected to the processor 300 and is used to receive UAV parameters, area coordinate information and control commands output by the processor 300, and to adjust the operation of the corresponding modules according to the above control commands. The area image acquisition module 720 includes an infrared camera and a hyperspectral camera. The infrared camera is used to acquire image data and temperature information of the area to be inspected, and can detect abnormal temperature of photovoltaic panels. The temperature measurement range is set to -20~150℃. The hyperspectral camera is used to acquire the material status of photovoltaic panels and water quality parameters. It can also be equipped with a visible light camera to record the panoramic view of the scene. The visible light image is used to analyze the changes in the reflectivity of the photovoltaic panel surface, calculate and evaluate dust accumulation, and analyze the distribution and activity intensity of fish through image sequence analysis. GPS tracking and positioning module 730 is used to track, locate and record the position of several UAVs in real time during flight, and correspond to the boundary position coordinates of the area to be inspected in memory 400. The obstacle avoidance and ranging module 740 is used to identify obstacles on the flight path and automatically adjust the flight path of the drone to maintain a safe distance between the drone and obstacles during flight. It can use ultrasonic or lidar to automatically avoid obstacles such as utility poles and trees. The communication module 750 is used to enable real-time communication between the ground control platform and the UAV, and to transmit the data received by the UAV control terminal 710 to the data analysis unit 200. The communication module 750 adopts 4G / 5G wireless communication technology or LoRa long-distance communication protocol to ensure real-time data transmission between the ground control platform and the UAV equipment 700. The UAV control terminal 710, the area image acquisition module 720, the GPS tracking and positioning module 730, the obstacle avoidance and ranging module 740, and the communication module 750 communicate with each other via a network. The UAV equipment 700 can take advantage of low-altitude flight to quickly reach areas that are difficult to be covered by the ground monitoring station 600, and obtain higher resolution image data. It can take pictures of the scene from all directions and multiple angles according to the preset flight route and time interval, making up for the limited field of view of the ground monitoring station 600. Specifically, the ground control platform generates task instructions based on inspection needs (such as daily inspections, abnormal re-inspections, and inspections after extreme weather), which are transmitted to the UAV control terminal 710 via the communication module 750. Parameters are configured, including flight altitude, airspeed, and camera activation. During the autonomous inspection phase, the UAV flies automatically along a preset route (such as a zigzag pattern in photovoltaic areas or a grid pattern in aquaculture areas). The GPS tracking and positioning module 730 corrects the position in real time, and the obstacle avoidance and ranging module 740 dynamically adjusts the path to avoid obstacles such as power lines and trees. The infrared camera captures temperature distribution, the hyperspectral camera detects panel aging and dust obstruction, and the visible light camera records fish distribution. After collecting the data, it is transmitted back to the ground control platform for subsequent analysis and operation. The UAV's path planning is based on coverage and priority of key areas. Specifically, the system also includes a user terminal, which is used to receive early warning information and view relevant data. The ground control platform, monitoring devices and user terminals communicate with each other through the network. The relevant data includes historical inspection data, abnormal data and prediction results.

[0025] In use, the area division unit 100 acquires environmental data of the area to be inspected in the solar-aquaculture complementary scenario, establishes an area coordinate system through the plot division module 110, and divides the area to be inspected into a photovoltaic area and an aquaculture area, each of which is further divided into several sub-areas. The area marking module 120 records the location and extent of the sub-areas and marks anomalies. The path planning module 130 plans the drone aerial photography route. The image monitor 610 of the ground monitoring station 600 takes images at regular intervals. The temperature and humidity sensor 620, water quality sensor 630, and meteorological sensor 640 collect corresponding environmental data and transmit them to the data analysis unit 200. Under the control of the drone processor, the drone equipment 700 flies according to the planned route, and the infrared camera of the area image acquisition module 720... The system includes a hyperspectral camera to collect images, temperature, photovoltaic panel material status, and water quality parameters; a GPS tracking and positioning module 730 for real-time positioning; an obstacle avoidance and ranging module 740 to ensure flight safety; a communication module 750 for data transmission; a data preprocessing module 210 in the data analysis unit 200 to process the collected data; a feature extraction module 220 to extract key features; a model prediction module 230 to predict the life cycle of the photovoltaic panel based on the prediction model; an early warning module 240 to compare data to identify anomalies and transmit them to the ground control platform; a result output module 250 to present the processing results; a memory 400 to store various types of data; a display panel 500 to display the surface condition and data information of the photovoltaic panel in real time; and a user terminal to receive early warning information and view relevant data. In summary, this system achieves comprehensive inspection of the solar-aquaculture hybrid scenario through the collaborative work of the ground control platform and monitoring devices. The area division unit 100 in the ground control platform rationally divides the inspection area and plans the path. The ground monitoring station collects images and environmental parameters of specific areas in real time through various sensors, while the UAV equipment collects data from multiple angles in the air. The data collected by the two complement each other to ensure the comprehensiveness of the data. The collected data is transmitted to the ground control platform through the communication module 750. The data analysis unit 200 performs preprocessing, feature extraction, and model prediction on the data to determine whether there are any abnormalities in the status of the photovoltaic panels, the environment of the aquaculture area, etc. If any abnormalities are found, an early warning is issued in a timely manner, so that users can understand the operation status of the scenario in real time and take corresponding measures in a timely manner.

[0026] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A fishlight complementary scene unmanned aerial vehicle inspection system, characterized in that, The ground control platform and the monitoring device are in communication with each other through a network, the ground control platform is used for periodically collecting and identifying image information and environmental data of a fishlight complementary scene, and performing processing, analysis and early warning operations on the image information and the environmental data, the monitoring device is used for monitoring the image and the environmental data of the fishlight complementary scene in an all-round way, and the ground control platform comprises: a region division unit (100) configured to acquire environmental data of a to-be-inspected region of the fishlight complementary scene, and divide the to-be-inspected region into a plurality of regions, the to-be-inspected region comprising a photovoltaic region and a breeding region; a data analysis unit (200) configured to analyze collected data from the monitoring device, and perform real-time update prediction on a service life of the photovoltaic panel based on the collected data; a processor (300) configured to coordinate data interaction between modules; a storage (400) configured to collect, arrange and store various data from the Internet, the monitoring device and the data analysis unit (200); a display panel (500) configured to display a surface condition of the photovoltaic panel and data information in real time; the monitoring device comprises: a ground monitoring station (600) configured to be equipped with a plurality of types of sensors, and acquire images and environmental parameters of a specific region in real time; an unmanned aerial vehicle device (700) configured to acquire image data of a surface of the photovoltaic panel through multi-angle shooting, and acquire overall environmental data in a use process of the fishlight complementary scene in real time. 2.The fish and light complementary scene unmanned aerial vehicle inspection system according to claim 1, characterized in that, the ground monitoring station (600) comprises: an image monitor (610) configured to shoot images of the surface of the photovoltaic panel and a surrounding area of the breeding region at regular time intervals, and transmit the acquired image information to the data analysis unit (200); a temperature and humidity sensor (620) configured to acquire environmental temperature and humidity of the fishlight complementary scene in real time; a water quality sensor (630) configured to detect water quality of the breeding region; a meteorological sensor (640) configured to collect, in real time, data of air temperature and climate, illumination time and intensity, wind speed, wind direction and rainfall in the fishlight complementary scene, the meteorological sensor (640) comprising an illumination intensity sensor, a wind speed and direction sensor and a rainfall sensor. 3.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 1, characterized in that, the unmanned aerial vehicle device (700) comprises a plurality of unmanned aerial vehicles and an unmanned aerial vehicle processor fixedly installed in the unmanned aerial vehicles, and the unmanned aerial vehicle processor comprises: an unmanned aerial vehicle control terminal (710) in electrical connection with the processor (300), configured to receive unmanned aerial vehicle parameters, region coordinate information and control instructions output by the processor (300), and perform operation adjustment on corresponding modules according to the control instructions; a region image acquisition module (720) comprising an infrared camera and a hyperspectral camera, the infrared camera being configured to acquire image data and temperature information of the to-be-inspected region, and the hyperspectral camera being configured to acquire a material state of the photovoltaic panel and water quality parameters; a GPS tracking and positioning module (730) configured to track and record positions of the plurality of unmanned aerial vehicles in real time during flight, and correspond to boundary position coordinates of the to-be-inspected region in the storage (400). The obstacle avoidance and distance measurement module (740) is configured to identify obstacles on the flight path and automatically adjust the flight path of the UAV, so that the UAV maintains a safe distance from the obstacles during flight. The communication module (750) is configured to realize real-time contact between the ground control platform and the UAV, and to transmit data received by the UAV control terminal (710) to the data analysis unit (200). The UAV control terminal (710), the regional image acquisition module (720), the GPS tracking and positioning module (730), the obstacle avoidance and distance measurement module (740), and the communication module (750) communicate with each other through a network. 4.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 1, characterized in that, The regional division unit (100) comprises: The plot division module (110) is configured to import regional drawings and field measurement data, establish a regional coordinate system, divide the fish-light complementary scene into a photovoltaic area and a breeding area, and divide the photovoltaic area and the breeding area into a plurality of sub-areas, and determine the regional coordinate information. The regional marking module (120) is configured to record the position and range of different sub-areas in the regional coordinate system, and display whether the corresponding area has a regional abnormal phenomenon through different colors. The path planning module (130) is configured to plan a plurality of UAV aerial photography routes based on the sub-area division results, GPS positioning data, photovoltaic panel size and installation angle, and distribution positions of the ground monitoring stations (600). 5.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 1, characterized in that, The data analysis unit (200) comprises: The data preprocessing module (210) is configured to perform preprocessing operations on the collected raw data, including cleaning, denoising, format conversion, and standardization processing. The feature extraction module (220) is configured to extract key features related to the photovoltaic panel state, the breeding area environment, and the overall scene from the preprocessed data. The model prediction module (230) is configured to construct a prediction model based on the photovoltaic panel life cycle data, historical environmental parameters, and operating data stored in the memory (400), and to real-time monitor and output the operation status and prediction results of the prediction model. The warning module (240) is configured to compare the simulation data generated by the prediction model with the real-time monitoring data to determine whether an abnormality occurs, and to transmit the abnormal data to the ground control platform if an abnormality occurs. The result output module (250) is configured to output the results processed by the data analysis unit (200) in the form of charts and reports, including photovoltaic panel life cycle prediction results, abnormality analysis, and environmental parameter statistical data. 6.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 3, characterized in that, The UAV flies according to the flight path and speed preset by the path planning module (130) under the control of the UAV processor, and the image data collected by the regional image acquisition module (720) and the image data collected by the ground monitoring station (600) form a complement. 7.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 5, characterized in that, The data preprocessing module (210) is further configured to convert the image data collected by the regional image acquisition module (720) into photo data with a time label, and to merge the photo data with the GPS positioning data recorded by the GPS tracking and positioning module (730) at the same time. 8.The fish-light complementary scene unmanned aerial vehicle inspection system of claim 5, wherein, The feature extraction module (220) extracts features including photovoltaic panel surface anomaly, water quality anomaly, fish activity anomaly, equipment failure, and stranger appearance, and the photovoltaic panel surface anomaly includes damage, hot spot, dust blockage, and support corrosion. 9.The fish-light complementary scene unmanned aerial vehicle inspection system of claim 5, wherein, The prediction model adopts a time series prediction model based on machine learning. 10.The fish-light complementary scene unmanned aerial vehicle inspection system according to claim 1, characterized in that, A user terminal is further included for receiving early warning information and viewing related data, the ground control platform, the monitoring device, and the user terminal communicate with each other through a network, and the related data includes historical inspection data, abnormal data, and prediction results.

Images